Issue Date: 6-7 March 2010rnrntOn page(s): rnt56rnttrn- 59rnrnrnLocation: Wuhan, ChinarnrnPrint ISBN: 978-1-4244-6388-6rnrnrnrnttrnDigital Object Identifier: href=''http://dx.doi.org/10.1109/ETCS.2010.160'' target=''_blank''>10.1109/ETCS.2010.160 rnrnDate of Current Version: trnrnt2010-05-06 14:33:52.0rnrnt rntt class="body-text">rntname="Abstract">>Abstractrn>In this paper, we propose a new framework in pedestrian detection using a two-step classification algorithm, which is a ȁC;
coarse to fineȁD;
course. The framework consists of a full-body detection (FBD) step and a head-shoulder detection (HSD) step. The FBD step uses fusion of Haar-like and HOG features to get better performance, and the HSD step utilizes edgelet;
Genetic Multiple Kernel; Relevance vector regression; genetic programming;
机译:基于跟踪逐架构的多行人跟踪调查
机译:基于磁区域回归的行人检测框架
机译:一种基于单深度图像的新型低虚警率行人检测框架
机译:基于新的两步框架的行人检测
机译:基于DDDAS的多尺度框架,用于行人行为建模以及与驾驶员的交互
机译:用于多个行人交叉检测的粗略框架
机译:自动缩放基于CNN的户外监视视频实时步行检测框架